[129502] trunk/dports/security

Joshua Root jmr at macports.org
Sun Dec 14 19:00:47 PST 2014


On 2014-12-15 13:48 , Kurt Hindenburg wrote:
> 
> On Dec 14, 2014, at 8:58 PM, Ryan Schmidt <ryandesign at macports.org> wrote:
> 
>>
>>> On Dec 14, 2014, at 11:59 AM, khindenburg at macports.org wrote:
>>
>> This doesn't look sufficient. If pinentry support is automatically included simply because pinentry is installed, then the main (non-pinentry) port needs to be fixed not to use pinentry even if it is already installed.
>>
>> Moreover, it seems strange to implement this as a subport instead of a variant. Unless a subport is a -devel subport, it's not supposed to conflict with the main port. Variants are typically used to enable optional features; pinentry support thus seems like a natural fit for a variant. What is the implication of having lastpass-cli built with pinentry support? I'm not really clear on what pinentry is.
> 
> My fault - I was under the impression we didn’t want to use variants anymore.  I’ll correct shortly.  pinentry is gnupg’s way to enter pin/password via dialogs.  

We don't want to use variants for things that other ports need to depend
on. Variants are fine when there are no dependents or they're not
affected by the chosen variants changing.

- Josh


More information about the macports-dev mailing list